More and more high-throughput data play a key role in medicine – both in the research of the pathophysiologic processes at molecular levels and the decision process in therapy. A personalised therapy strategy usually requires the analysis of patient related high-throughput data (e.g. genome, exome, metagenome). Of same importance are transcriptome, proteome and metabolome data, which are essential biomarkers for diagnosis and stratification of patient sub-groups. Whereas the analysis of classical low-throughput data is comparatively easy, high-throughput data with huge data volume and high complexity make greater demands on the analysis. Efficient algorithms for information extraction and statistically robust interpretation of the results are inevitable in this field. The department Complex Diagnostics investigates innovative methods for automated analysis of high-throughput data that produce reliable data in short time. Besides innovative algorithms the establishment of a high-performance infrastructure for automated data processing is one of our main focuses. By integration of these newly developed algorithms into well-established workflows and by direct link-ups to clinical information systems, translation of new results from research into clinics can be achieved very quickly. One major aspect of the acceptance of such new methods in clinical routines is a user-friendly interface with an intuitive visualization concept. Therefore cognition scientists and media/information scientists work together in this field of research to make the data sets intuitively and interactively understandable to achieve an early integration of high-throughput data into the clinical routine.